Abstract: In order to evaluate the sensory quality indexes affecting consumer likability on pipe tobacco and to understand consumer preference for pipe tobacco products, online data on pipe tobacco products were obtained by a web crawler based on Python programming language. The data were analyzed with orthogonal partial least squares discriminant analysis and stepwise regression analysis. The results showed that consumer likability for original-flavor pipe tobacco products was mainly based on two pipe smoke sensory indexes: gradient variation and texture thickness; between the two indexes, gradient variation was more important than texture thickness. The consumer likability for flavored pipe tobacco products could be summarized mainly on the two sensory indexes: gradient variation and environmental flavor note, with gradient variation being more important than the environmental note.